Acute myocardial infarction remains common with a reported annual incidence of 1.1 million cases in the United States alone (Antman, E. M., Braunwald, E., Acute myocardial infarction, in Principles of Internal Medicine, 15th Ed., Braunwald, E. et al., Eds., New York: McGraw-Hill (2001)). Preclinical and clinical data demonstrate that following a myocardial infarction, the acute loss of myocardial muscle cells and the accompanying peri-infarct zone hypo-perfusion result in a cascade of events causing an immediate diminution of cardiac function, with the potential for long term persistence. The extent of myocardial cell loss is dependent on the duration of coronary artery occlusion, existing collateral coronary circulation and the condition of the cardiac microvasculature. Paul et al., Am. Heart J. 131: 710-15 (1996); Pfeffer, M. A., Braunwald, E., Circulation 81: 1161-72 (1990); Sheiban, I.e. al., J. Am. Coll. Cardiol. 38: 464-71 (2001); Braunwald E., Bristow, M. R., Circulation 102: IV-14-23 (2000); Rich et al., Am. J. Med. 92:7-13 (1992); Ren et al., J. Histochem. Cytochem. 49: 71-79 (2002); Hirai, T. et al., Circulation 79: 791-96 (1989); Ejiri, M. et al., J. Cardiology 20: 31-37 (1990). Because myocardial cells have virtually no ability to regenerate, myocardial infarction leads to permanent cardiac dysfunction due to contractile-muscle cell loss and replacement with nonfunctioning fibrotic scarring. Frangogiannis, N. G., et al., Cardiovascular Res. 53(1): 31-47 (2002). Moreover, compensatory hypertrophy of viable cardiac muscle leads to microvascular insufficiency that results in further demise in cardiac function by causing myocardial muscle hibernation and apoptosis of hypertrophied myocytes in the peri-infarct zone.
Among survivors of myocardial infarction, residual cardiac function is influenced most by the extent of ventricular remodeling (meaning changes in size, shape, and function, typically a decline in function, of the heart after injury). Alterations in ventricular topography (meaning the shape, configuration, or morphology of a ventricle) occur in both infarcted and healthy cardiac tissue after myocardial infarction. Pfeffer, M. A., Braunwald, E., Circulation 81: 1161-72 (1990). Ventricular dilatation (meaning a stretching, enlarging or spreading out of the ventricle) causes a decrease in global cardiac function and is affected most by the infarct size, infarct healing and ventricular wall stresses. Recent efforts to minimize remodeling have been successful by limiting infarct size through rapid reperfusion (meaning restoration of blood flow) using thrombolytic agents and mechanical interventions, including, but not limited to, placement of a stent, along with reducing ventricular wall stresses by judicious use of pre-load therapies and proper after-load management. Id. Regardless of these interventions, a substantial percentage of patients experience clinically relevant and long-term cardiac dysfunction after myocardial infarction. Sheiban, I. et al., J. Am. Coll. Cardiol. 38: 464-71 (2001). Despite revascularization of the infarct related artery circulation and appropriate medical management to minimize ventricular wall stresses, a significant percentage of patients experience ventricular remodeling, permanent cardiac dysfunction, and consequently remain at an increased lifetime risk of experiencing adverse cardiac events, including death. Paul et al., Am. Heart J. 131: 710-15 (1996); Pfeffer, M. A., Braunwald, E., Circulation 81: 1161-72 (1990).
At the cellular level, immediately following a myocardial infarction, transient generalized cardiac dysfunction uniformly occurs. In the setting of a brief (i.e., lasting three minutes to five minutes) coronary artery occlusion, energy metabolism is impaired, leading to demonstrable cardiac muscle dysfunction that can persist for up to 48 hours despite immediate reperfusion. This so-called “stunned myocardium phenomenon” occurs subsequent to or after reperfusion and is thought to be a result of reactive oxygen species. The process is transient and is not associated with an inflammatory response. Frangogiannis, N. G., et al., Cardiovascular Res. 53(1): 31-47 (2002). After successful revascularization, significant recovery from stunning occurs within three to four days, although complete recovery may take much longer. Boli, R., Prog. Cardiovascular Disease 40(6): 477-515 (1998); Sakata, K. et al., Ann. Nucleic Med. 8: 153-57 (1994); Wollert, K. C. et al., Lancet 364: 141-48 (2004).
Coronary artery occlusion of more significant duration, i.e., lasting more than five minutes, leads to myocardial ischemia (i.e., an insufficient blood flow to the heart's muscle mass) and is associated with a significant inflammatory response that begins immediately after reperfusion and can last for up to several weeks. Frangogiannis, N. G., et al., Cardiovascular Res. 53(1): 31-47 (2002); Frangogiannis, N. G. et al., Circulation 98: 687-798 (1998).
The inflammatory process following reperfusion is complex. Initially it contributes to myocardial damage but later leads to healing and scar formation. This complex process appears to occur in two phases. In the first so-called “hot” phase (within the first five days), reactive oxygen species (in the ischemic myocardial tissue) and complement activation generate a signal chemotactic for leukocytes (chemotaxis is the directed motion of a motile cell, organism or part towards environmental conditions it deems attractive and/or away from surroundings it finds repellent) and initiate a cytokine cascade. Lefer, D. J., Granger, D. N., Am. J. Med. 4:315-23 (2000); Frangogiannis, N. G., et al., Circulation 7:699-710 (1998). Mast cell degranulation, tumor necrosis factor alpha (TNF-α) release, and increased interleukin-6 (IL-6), intercellular adhesion molecule 1 (“ICAM-1” or CD-54, a receptor typically expressed on endothelial cells and cells of the immune system), selectin (L, E and P) and integrin (CD11a, CD11b and CD18) expression all appear to contribute to neutrophil accumulation and degranulation in ischemic myocardium. Frangogiannis, N. G. et al., Circulation 7: 699-710 (1998), Kurrelmeyer, K. M, et al., Proc. Nat'l Acad. Sci. 10: 5456-61 (2000); Lasky, L. A., Science 258: 964-69 (1992); Ma, X. L., et al., Circulation 88(2): 649-58 (1993); Simpson, P. J. et al., J. Clin. Invest. 2: 624-29 (1998). Neutrophils contribute significantly to myocardial cell damage and death through microvascular obstruction and activation of neutrophil respiratory burst pathways after ligand-specific adhesion to cardiac myocytes. Entman, M. L., et al., J. Clin. Invest. 4: 1335-45 (1992). During the “hot” phase, angiogenesis is inhibited due to the release of angiostatic substances, including interferon gamma-inducible protein (IP 10). Frangogiannis, N. G., et al., FASEB J. 15: 1428-30 (2001).
In the second phase, the cardiac repair process begins (about day 6 to about day 14), which eventually leads to scar formation (about day 14 to about day 21) and subsequent ventricular remodeling (about day 21 to about day 90). Soon after reperfusion, monocytes infiltrate the infarcted myocardium. Attracted by complement (C5a), transforming growth factor B1 (“TGF-B1”) and monocyte chemotactic protein 1 (“MCP-1”), monocytes differentiate into macrophages that initiate the healing process by scavenging dead tissue, regulating extracellular matrix metabolism, and inducing fibroblast proliferation. Birdshall, H. H., et al., Circulation 3: 684-92 (1997). Secretion of interleukin 10 (IL-10) by infiltrating lymphocytes also promotes healing by down-regulating inflammatory cytokines and influencing tissue remodeling. Frangogiannis, N. G. et al., J. Immunol. 5:2798-2808 (2000). Mast cells also appear to be involved in the later stages of myocardial repair by participating in the formation of fibrotic scar tissue. Stem Cell Factor (SCF) is a potent attractor of mast cells. SCF mRNA has been shown to be up-regulated in ischemic myocardial segments in a canine model of myocardial infarction and thus may contribute to mast cell accumulation at ischemic myocardial sites. Frangogiannis, N. G. et al., Circulation 98: 687-798 (1998). Mast cell products (including TGF-B, basic fibroblast growth factor (bFGF), vascular endothelial growth factor (VEGF) and gelatinases A and B) induce fibroblast proliferation, influence extracellular matrix metabolism, and induce angiogenesis. Fang, K. C., et al., J. Immunol. 162: 5528-35 (1999); Takeshi, S., et al., Cardiology 93: 168-74 (2000).
Following a myocardial infarction, neoangiogenesis occurs after the “hot” phase of the inflammatory process subsides (about day 5) coincident with rising levels of VEGF (VEGF peaks at about day 7 and gradually subsides to baseline at about day 14 to about day 21). During this phase of the healing process, endothelial precursor cells (EPCs) are mobilized and recruited to the infarct site. Shinitani, S., et al., Circulation 103: 2776-79 (2001). Without being limited by theory, it has been suggested that the chemokine stromal cell derived factor-1 (SDF-1), which is the ligand for the CXCR-4 chemokine receptor expressed by CD34+ cells, also plays a role in homing of cells to areas of ischemic damage. Ceredini, D. J., et al., Nature Medicine 10: 858-63 (2004); Askari, A., et al., Lancet 362: 697-703 (2003); Yamaguchi, J. et al., Circulation 107: 1322-34 (2003). While it is known that SDF-1 plays a role in hematopoiesis and is involved in migration, homing and survival of hematopoietic progenitors, and while SDF-1 has been implicated in ischemic neovascularization in vivo by augmenting EPC recruitment to ischemic sites (Yamaguchi et al. Circulation 107:1322-1328 (2003), SDF-1's role in neoangiogenesis is not certain. There is suggestive evidence implicating SDF-1. For example, SDF-1 gene expression is upregulated during hypoxia, a deficiency of oxygen in the tissues, by hypoxia inducible factor-1. Furthermore, CD34+ cells are capable of homing to areas of ischemia, rich in SDF-1, including infarcted myocardium. Askari et al., Lancet 362: 697-703 (2003). Moreover, virtually all CD34+ CXCR-4+ cells co-express VEGF-2 and therefore migrate in response to VEGF as well as SDF-1. Peichev M., et al., Blood 95: 952-58 (2000). CD34+CXCR-4+ VEGF-1 cells, once recruited, are capable of contributing to neoangiogenesis. Yamaguchi, J. et al., Circulation 107: 1322-34 (2003).
To date, no ideal therapy exists for preventing the long term adverse consequences of vascular insufficiency, particularly the significant vascular insufficiency that results in a myocardial infarction. While large vessel revascularization (meaning the successful placement of a stent) seems promising, studies to date have shown such applications to be insufficient in addressing increased demands posed by compensatory myocardial hypertrophy. As a result, infarct extension and fibrous replacement commonly occur, regardless of large vessel revascularization, appropriate medical management of ventricular wall stresses, and potential natural, albeit suboptimal, CD34+ cell-mediated neoangiogenesis (one of the theories relating to the underlying cause of myocardial infarction is that the ability to mobilize these cells may be biologically limited).
Intense interest has developed in evaluating the ability of endothelial and myocardial precursor cells to limit damage to the myocardium after infarction and to limit or prevent ventricular remodeling. Significant preclinical data and some clinical data demonstrate the safety and potential of cell therapy using a variety of cell precursors (particularly hematopoietic cells) to contribute to neoangiogenesis, limited cardiac myogenesis (principally by fusion), and muscle preservation in the myocardial infarct zone. See, e.g., Jackson, et al., J. Clin. Invest. 107: 1395-1402 (2001); Edelberg, J. M., et al., Cir. Res. 90: e89-e93 (2002); Schichinger, V. et al., New Engl. J. Med. 355 (12): 1210-21 (2006) (using bone marrow-derived progenitor cells); Assmus, B. et al., New Engl. J. Med. 355 (12) 1222-32 (2006) (using bone marrow-derived progenitor cells), but see Lunde, K. et al., New Eng. J. Med. 355 (12): 1199-209 (2006) (using bone marrow-derived progenitor cells).
Bone marrow consists of a variety of precursor and mature cell types, including hematopoietic cells (the precursors of mature blood cells) and stromal cells (the precursors of a broad spectrum of connective tissue cells), both of which appear to be capable of differentiating into other cell types. Wang, J. S. et al., J. Thorac. Cardiovasc. Surg. 122: 699-705 (2001); Tomita, S. et al., Circulation 100 (Suppl. II): 247-256 (1999); Saito, T. et al., Tissue Eng. 1: 327-43 (1995). Unmodified (i.e., not fractionated) marrow or blood-derived cells have been used in several clinical studies, for example, Hamano, K. et al., Japan Cir. J. 65: 845-47 (2001); Strauer, B. E., et al., Circulation 106: 1913-18 (2002); Assmus, et al., Circulation 106: 3009-3017 (2002); Dobert, N. et al., Eur. J. Nucl. Med. Mol. Imaging, 8: 1146-51 (2004); Wollert, K. C. et al., Lancet 364: 141-48 (2004). Since the mononuclear fraction of bone marrow contains stromal cells, hematopoietic precursors, and endothelial precursors, the relative contribution of each of these populations to the observed effects, if any, remains unknown.
CD34 is a hematopoietic stem cell antigen selectively expressed on hematopoietic stem and progenitor cells derived from human bone marrow, blood and fetal liver. Yin et al., Blood 90: 5002-5012 (1997); Miaglia, S. et al., Blood 90: 5013-21 (1997). Cells that express CD34 are termed CD34+. Stromal cells do not express CD34 and are therefore termed CD34−. CD34+ cells isolated from human blood may be capable of differentiating into cardiomyocytes, endothelial cells, and smooth muscle cells in vivo. See Yeh, et al., Circulation 108: 2070-73 (2003). CD34+ cells represent approximately 1% of bone marrow derived nucleated cells; CD34 antigen also is expressed by immature endothelial cell precursors; mature endothelial cells do not express CD34+. Peichev, M. et al., Blood 95: 952-58 (2000). In vitro, CD34+ cells derived from adult bone marrow give rise to a majority of the granulocyte/macrophage progenitor cells (CFU-GM), some colony-forming units-mixed (CFU-Mix) and a minor population of primitive erythroid progenitor cells (burst forming units, erythrocytes or BFU-E). Yeh, et al., Circulation 108: 2070-73 (2003). CD34+ cells also may have the potential to differentiate into or to contribute to the development of new myocardial muscle, albeit at low frequency.
Techniques have been developed using immunomagnetic bead separation to isolate a highly purified and viable population of CD34+ cells from bone narrow mononuclear cells. See U.S. Pat. Nos. 5,536,475, 5,035,994, 5,130,144, 4,965,205, the contents of each of which is incorporated herein by reference. Two clinical studies support the clinical application of bone marrow derived CD34+ cells after myocardial infarction. See C. Stamm, et al., Lancet 361: 45-46 (2003); Herenstein, B. et al., Blood Supplement, Abs. 2696 (2004).
To date, however, no ideal therapy exists for preventing the long-term adverse consequences of vascular insufficiency, particularly vascular insufficiency that produces myocardial infarction.
The present invention addresses the question of whether a composition comprising a chemotactic hematopoietic stem cell product comprising an enriched population of isolated CD34+ cells containing a subpopulation of cells having chemotactic activity can be manufactured, remain stable for a commercially viable period, and be delivered to a subject in need thereof so that potent cells can home to and repair sites of vascular insufficiency, including infarcted myocardium.